Data have been collected for two experiments concerning influences of alcohol and tract configuration (lane width in the first, and curve characteristics in the second study) upon driving behavior as measured by control use and vehicle performance. A standard domestic station wagon was instrumented to record analog information; i.e., steering wheel position, speed, accelerator pedal position, brake pedal position, brake pressure, and longitudinal and lateral acceleration. All driving was done at night on a closed course isolated from traffic. In the first study, a partially nested repeated measures design was employed, with the 18 subjects tested on four nonconsecutive nights. The first night was for practice to reduce learning and transfer effects. On the remaining three nights, the subject was tested following the consumption of beverages mixed to produce different target BACs (0, 50, 100 milligrams %). In the second study, a factorial design was employed, with the 12 subjects tested on five nonconsecutive nights. No practice night was given in order to examine the effects of alcohol upon learning. The BAC targets used were 0, 50, and 100 milligrams %, with the 0 and 100 milligrams % conditions administered twice. $ Computer programs for analysis have been completed, and data processing is in progress.